2000 character limit reached
Learning Sequence Attractors in Recurrent Networks with Hidden Neurons (2404.02729v1)
Published 3 Apr 2024 in cs.NE, cs.AI, and cs.LG
Abstract: The brain is targeted for processing temporal sequence information. It remains largely unclear how the brain learns to store and retrieve sequence memories. Here, we study how recurrent networks of binary neurons learn sequence attractors to store predefined pattern sequences and retrieve them robustly. We show that to store arbitrary pattern sequences, it is necessary for the network to include hidden neurons even though their role in displaying sequence memories is indirect. We develop a local learning algorithm to learn sequence attractors in the networks with hidden neurons. The algorithm is proven to converge and lead to sequence attractors. We demonstrate that the network model can store and retrieve sequences robustly on synthetic and real-world datasets. We hope that this study provides new insights in understanding sequence memory and temporal information processing in the brain.
- Amari, S.-I. (1972). Learning patterns and pattern sequences by self-organizing nets of threshold elements. IEEE Transactions on Computers.
- Amit, D. J. (1989). Modeling Brain Function: The World of Attractor Neural Networks. Cambridge University Press.
- Bengio, Y. (2014). How auto-encoders could provide credit assignment in deep networks via target propagation. arXiv preprint arXiv:1407.7906.
- Matching recall and storage in sequence learning with spiking neural networks. Journal of Neuroscience.
- Perceptron-like learning in time-summating neural networks. Journal of Physics A: Mathematical and General.
- Long sequence hopfield memory. Advances in Neural Information Processing Systems.
- Links between perceptrons, mlps and svms. International Conference on Machine learning.
- Hippocampal replay of extended experience. Neuron.
- Eichenbaum, H. (2014). Time cells in the hippocampus: a new dimension for mapping memories. Nature Reviews Neuroscience.
- Spike-time-dependent plasticity and heterosynaptic competition organize networks to produce long scale-free sequences of neural activity. Neuron.
- Reverse replay of behavioural sequences in hippocampal place cells during the awake state. Nature.
- Neuromodulated spike-timing-dependent plasticity, and theory of three-factor learning rules. Frontiers in Neural Circuits.
- Gardner, E. (1988). The space of interactions in neural network models. Journal of Physics A: Mathematical and General.
- Characteristics of sequential activity in networks with temporally asymmetric hebbian learning. Proceedings of the National Academy of Sciences.
- Hopfield, J. J. (1982). Neural networks and physical systems with emergent collective computational abilities. Proceedings of the National Academy of Sciences.
- The ou-isir gait database comprising the large population dataset and performance evaluation of gait recognition. IEEE Transactions on Information Forensics and Security.
- General sequential episodic memory model. International Conference on Machine Learning.
- Attractor and integrator networks in the brain. Nature Reviews Neuroscience.
- Kleinfeld, D. (1986). Sequential state generation by model neural networks. Proceedings of the National Academy of Sciences.
- Dense associative memory for pattern recognition. Advances in Neural Information Processing Systems.
- Learning with three factors: modulating hebbian plasticity with errors. Current Opinion in Neurobiology.
- Robust timing and motor patterns by taming chaos in recurrent neural networks. Nature Neuroscience.
- LeCun, Y. (1986). Learning process in an asymmetric threshold network. Disordered Systems and Biological Organization.
- LeCun, Y. (1987). Modeles connexionnistes de lapprentissage. PhD thesis, These de Doctorat, Universite Paris.
- Memory of sequential experience in the hippocampus during slow wave sleep. Neuron.
- Random synaptic feedback weights support error backpropagation for deep learning. Nature Communications.
- Backpropagation and the brain. Nature Reviews Neuroscience.
- Optimal degrees of synaptic connectivity. Neuron.
- Perceptrons. MIT Press.
- Exponentially long orbits in hopfield neural networks. Neural Computation.
- Replay and time compression of recurring spike sequences in the hippocampus. Journal of Neuroscience.
- Internally generated cell assembly sequences in the rat hippocampus. Science.
- Hippocampal place-cell sequences depict future paths to remembered goals. Nature.
- Autoassociative dynamics in the generation of sequences of hippocampal place cells. Science.
- Stimulus-driven and spontaneous dynamics in excitatory-inhibitory recurrent neural networks for sequence representation. Neural Computation.
- Recurrent network models of sequence generation and memory. Neuron.
- Temporal association in asymmetric neural networks. Physical Review Letters.
- Unsupervised learning of video representations using lstms. International Conference on Machine Learning.
- Sequential memory with temporal predictive coding. Advances in Neural Information Processing Systems.
- Spike-based bayesian-hebbian learning of temporal sequences. PLoS Computational Biology.
- Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology.
- Vershynin, R. (2018). High-Dimensional Probability: An Introduction with Applications in Data Science. Cambridge University Press.